Flight Booking System Microservices

Overview

This project is a Flight Booking System built using Java and Spring Boot Reactive WebFlux following a microservices architecture.
An existing monolithic application was split into independent services with separate databases, service discovery, centralized configuration, fault tolerance, and asynchronous communication.

System Architecture

View Summary Document

The diagram shows the API Gateway as a single entry point, independent core services with their own databases, service discovery using Eureka, centralized configuration, and asynchronous communication using Kafka.

---

Microservices

• API Gateway - works on 9000 port
• Flight Service - works on 8090 port
• Booking Service - works on 8080
• Config Server - works on 8761 port

Flight Service

• Handles all flight related operations such as adding airlines, adding flights, and searching flights.
• Implemented using Reactive WebFlux and MongoDB with database name flight_db.
• Unit tested with JaCoCo coverage of 96 percent.

Booking Service

• Manages booking operations including booking a flight, cancelling tickets, fetching booking details by email, and retrieving tickets using PNR.
• Implemented using Reactive WebFlux and communicates with Flight Service using OpenFeign.
• A Circuit Breaker is applied to handle service failures.
• Uses MongoDB with database name booking_db.
• Unit tested with JaCoCo coverage of 92 percent.

Notification Service

Consumes Kafka events and sends booking confirmation emails.
Email sending is asynchronous and decoupled from the booking flow.

---

API Gateway Endpoints

All client requests are routed through the API Gateway.

HTTP Method	Endpoint	Description	Forwarded To
GET	/flights/search	Search available flights	Flight Service
POST	/flights	Add a new flight	Flight Service
POST	/airlines	Add airline details	Flight Service
POST	/bookings	Book a flight	Booking Service
DELETE	/bookings/{pnr}	Cancel ticket using PNR	Booking Service
GET	/bookings/email/{email}	Get booking details by email	Booking Service
GET	/bookings/pnr/{pnr}	Get ticket details using PNR	Booking Service

---

Supporting Components

• API Gateway used as a single entry point for all client requests
• Eureka Server for service discovery
• Config Server for centralized configuration management
• Apache Kafka as a message broker for email notifications
• Proper validations and exception handling have been implemented in each service

---

Testing and Code Quality

Unit and Load Testing

• Unit testing completed for core services. Flight Service - 96 percent and Booking Service - 92 percent
• Overall test coverage is 93.38 percent
• JMeter load testing performed using CLI with 20, 50, and 100 threads

SonarQube Results

• Zero bugs
• Zero vulnerabilities
• Zero code duplications
• High maintainability rating

---